| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
INSERM, U-515, Croissance, Differenciation et Processus Tumoraux, Hôpital Saint-Antoine (C.M., C.G., M.L., Y.L.B.), 75571 Paris, France; Service d’Anatomo-Pathologie, Hôpital Cochin (A.L.), Assistance Publique-Hôpitaux de Paris, 75014 Paris, France; and Department of Anatomy, University of Cambridge (P.N.S.), Cambridge, United Kingdom
Address all correspondence and requests for reprints to: Dr. C. Martinerie, INSERM, U-515, Hôpital Saint-Antoine, 184 rue du Faubourg Saint-Antoine, 75571 Paris Cedex 12, France. E-mail: martiner{at}st-antoine.inserm.fr
Abstract
NOVH belongs to the CCN (CTGF/CYR61/NOV) family of proteins, some of which have chemotactic, mitogenic, adhesive, and angiogenic properties. Whereas ctgf and cyr61 are growth factor-inducible, immediate-early genes, nov is expressed in growth-arrested or quiescent cells. As nov expression has been shown to be altered in both avian and human nephroblastomas and to be a target of WT1 regulation, NOV may play important roles in normal nephrogenesis and the development of Wilms’ tumors.
The aim of this study was to determine whether changes in novH expression were associated with tumorigenesis in tissues other than those of the kidney. We showed by Northern blotting and immunohistochemistry that among human adult endocrine tissues, the adrenal gland is a major site of novH expression, and that in adult and fetal adrenal tissue, novH is primarily expressed in the adrenal cortex. Studies with 12 benign and 18 malignant adrenocortical tumors revealed that the levels of novH mRNA and protein decreased significantly (P < 0.004) with progression of adrenocortical tumors from a benign to a malignant state. Although the localization of NOVH did not change, the N-glycosylation profile of benign and malignant tumors differed considerably from that of normal adrenocortical tissue, and these differences may affect the biochemical properties of the molecule. The properties of NOVH here provide the first evidence that this member of the CCN family could be involved in adrenocortical tumor development.
ADRENOCORTICAL CARCINOMAS are rare tumors with a poor prognosis (1, 2). Their pathogenesis is not completely understood, but evidence is accumulating that the insulin-like growth factor (IGF) system plays a major role in adrenocortical tumorigenesis (2, 3, 4). Alterations to the IGF system have been demonstrated in malignant adrenocortical tumors. The changes observed include imprinting mistakes of the 11p15 region, overexpression of the IGF-II gene and of its receptor IGF-I receptor and high levels of IGF-binding protein-2 (IGFBP-2) (2, 3, 5, 6). In addition, IGF-II has been shown to be involved in the auto/paracrine proliferation of H295R cells, an in vitro model for adrenocortical carcinoma (4). Imprinting mistakes of the 11p15 region are also responsible for the loss of expression of CK1 p57Kip2 and, consequently, for overactivity of G1/S phase cyclin-CDK complexes (7).
Dysregulation of imprinted growth regulatory genes within the 11p15 region is also involved in the Beckwith-Wiedmann syndrome, an overgrowth syndrome predisposing patients to various tumors, including nephroblastoma (Wilms’ tumor) and adrenocortical carcinoma (8). The avian nephroblastoma induced by the myeloblastosis-associated virus 1-N constitutes a unique animal model of Wilms’ tumor (9). Molecular cloning of myeloblastosis-associated virus 1-N integration sites in avian nephroblastoma resulted in identification of the nephroblastoma overexpressed (nov) gene (10, 11). nov belongs to the recently discovered CCN [ctgf (12), cyr 61, (13), and nov] family of genes (14), which also includes elm1/wisp1 (15, 16), r-cop1/wisp2 (15, 17, 18), and wisp3 (15). This family of genes has been previously included in the IGFBP superfamily (19). nov has been cloned from chicken, human, mouse, and Xenopus (10, 20, 21, 22) and is well conserved throughout evolution. The biological properties of this family of genes include the regulation of cell proliferation, chemotaxis, angiogenic and adhesive activities, and extracellular matrix formation. In vivo, the CCN family appears to be involved in both normal processes, such as implantation, placentation, embryogenesis differentiation, and development, and in pathological situations, including wound healing, fibrotic disorders, and tumors (for a review, see Ref. 23).
nov expression is altered in both avian and human nephroblastomas (10, 20, 24). In Wilms’ tumors nov expression is altered in the blastema and is associated with heterotypic blastemal differentiation (24). In addition, we showed that levels of nov and Wilms’ tumor suppressor gene (wt1) mRNA were inversely correlated in several Wilms’ tumors and that nov was down-regulated by WT1 proteins in ex vivo assays and was therefore a potential target for wt1 regulation (20, 25). High levels of IGF-II and IGF-I receptor have also been described in these tumors (26, 27, 28), and there is an inverse correlation between wt1 and IGF-I receptor (29). Moreover, WT1 proteins inhibit IGF-II transcription (30).
nov expression is not restricted to kidney and has been detected in other tissues, such as brain, muscle, cartilage, bone, and lung (10, 24, 31). It is also associated with the development of the central nervous system in humans (32). These observations raise the possibility that novH may be involved in diseases in organs other than the kidney. As Wilms’ and adrenocortical tumors have some physiopathological and molecular alterations in common (33), we investigated novH expression in the adrenal cortex. We found that among endocrine glands, this tissue was a major site of novH expression in adults and during embryogenesis and that quantitative and qualitative changes in novH expression correlated with the acquisition by adrenocortical tissue of a tumoral phenotype. Thus, alterations of novH expression may play a role in this tumorigenesis.
Subjects and Methods
Patients
Thirty patients with sporadic adrenocortical tumors, aged 16–79 yr, were included in this study. Hormonal status and stage of the tumor were assessed as previously described (34). Histological features, including, high mitotic rate, atypical mitoses, high nuclear grade, low percentage of clear cells, necrosis, diffuse architecture of tumor, capsular invasion, sinusoidal invasion, and venous invasion, were carefully investigated. Tumors with none of these histological features were classified as benign. Localized tumors with one to three of these histological features were classified as suspect. Tumors with more than three of these features or a history of metastasis or recurrence were classified as malignant (35).
Two groups of tumors were considered based firstly on pathological data
and secondly on 11p15 molecular abnormalities: group 1 (n = 12),
all benign tumors and suspect tumors with no 11p15 abnormalities; and
group 2 (n = 18), all malignant tumors and suspect tumors with
11p15 abnormalities. Pathological, hormonal and molecular data are
summarized in Table 1
. Patients were
numbered after entry in the study.
|
Tumor samples were obtained during surgery, carefully dissected by the pathologist, and immediately frozen and kept at -80 C.
Protein extractions
Frozen tissues (mean weight, 100–300 mg) were quickly homogenized on ice in 3 ml lysis buffer [50 mM HEPES (pH 7), 250 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40, and 1 mM dithiothreitol] containing proteases inhibitors (1 µg/ml aprotinin, 1 µg/ml leupeptin, and 50 µg/ml phenylmethylsulfonylfluoride) and phosphatase inhibitors (1 mM orthovanadate and 2 mM sodium pyrophosphate), using a Polytron (Brinkmann Instruments, Inc., Westbury NY). The homogenates were incubated for 1 h at 0 C and centrifuged at 20,000 x g for 15 min at 4 C. The supernatants were removed and frozen at -80 C. Small aliquots of the supernatants were used for protein determination (protein assays from Bio-Rad Laboratories, Inc., Richmond, CA).
Immunoblotting
Protein samples (40 µg) were subjected to 12% SDS-PAGE under reducing conditions and were transferred to polyvinylidene difluoride membranes (Hybond P, Amersham Pharmacia Biotech, Orsay, France) for immunological detection.
The K19M anti-NOVH polyclonal antibody has been described previously (24). A 1:500 dilution of K19M antibody was first incubated with the membrane for 1 h at 37 C. Immunoreactive proteins were detected by enhanced chemiluminescence (Amersham Pharmacia Biotech) according to the manufacturer’s instructions.
N-glycosidase F treatment
Proteins were precipitated from extracts of adrenal tissue (40 µg) or from conditioned medium (20 µl) from SF9 insect cells synthesizing NOVH protein, by incubation at 4 C in the presence of 0.02% sodium deoxycholate in 100 mM Tris-HCl, pH 8.5, and 20% trichloroacetic acid. The precipitated protein was pelleted by centrifugation at 20,000 x g for 10 min and was then washed twice with 500 µl acidified acetone (10 mM HCl) to extract and dissolve trichloroacetic acid and deoxycholate. Acetone was then eliminated with 500 µl diethyl ether, and the protein pellet was dried for 5 min at 37 C. Proteins were then dissolved and denatured in 100 µl N-glycosidase F buffer [50 mM sodium phosphate buffer (pH 7.2), 10 mM EDTA, 0.1% SDS, and 1% ß-mercaptoethanol]. Nonidet P-40 (1%) was added to neutralize SDS, and protein samples were treated with 5 U N-glycosidase F (1 U/µl; Roche, Meylan, France) for 16 h at 37 C. Protein samples incubated for the same time at 37 C without enzyme were used as controls.
Immunohistochemistry
Immunohistochemistry was performed on 4-µm, formalin- or Bouin-fixed, paraffin-embedded sections as previously described (36) using the NOVH–specific K19M antibody at a 1:250 dilution. The peroxidase reaction was developed for 5 min in diaminobenzidine solution (DAKO Corp., Glostrup, Denmark), and sections were counterstained with Mayer’s hematoxylin solution (Labonord, Villeneuve d’Asq, France), dehydrated, and mounted with Eukitt (Labonord). Controls were incubated without the primary antibody, or the K19M antibody was first incubated with 10 µg/ml NOVH-specific antigen. In neither control was any specific staining observed. The specificity of detection was also checked with affinity-purified K19M polyclonal antibody with similar results. Antichromogranin A monoclonal antibody (Biosoft,Varilhes, France) was used at a 1:100 dilution. Anti-PS100 polyclonal antibody (DAKO Corp.) was used at a 1:500 dilution. The alkaline phosphatase reaction was developed in Fast Red solution (DAKO Corp.). After hematoxylin-eosin-saffron staining or after incubation with antibodies, the tissues were examined by an experienced pathologist (A.L.) in adrenal tissues.
RNA extraction and Northern blotting
A Northern blot assay with 2 µg polyadenylated [poly(A)+] RNA from normal adult endocrine tissues was purchased from CLONTECH Laboratories, Inc. Total RNA was extracted from frozen adrenal tissues by the CsCl/guanidine isothiocyanate method (37). Samples of poly(A)+ or total RNA (10 µg) were loaded on a 1% agarose-2.2 mol/liter formaldehyde gel, subjected to electrophoresis, and transferred onto nylon membranes. The membranes were hybridized as previously described with the 1.9-kb EcoRI novH probe (20, 24) labeled by random hexamer priming (Amersham Pharmacia Biotech) using [32P]deoxy-CTP. The signal for novH was normalized using the intensity of signal for ß-actin or glyceraldehyde-3-phosphate dehydrogenase (GAPDH; CLONTECH Laboratories, Inc.). For Northern blot comparisons, total RNA extracted from HeLa cells (15 µg) was used as a reference sample.
Densitometry
Western immunoblots were analyzed by scanning with a GS700 imaging densitometer and processing with the Molecular Analyst data system (Bio-Rad Laboratories, Inc.).
Northern blots were analyzed with a Storm PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA).
Statistical analysis
Data are expressed as the mean ± SEM. The two groups of tumors were compared by Mann-Whitney’s U test for unpaired data using Statistica software (Stat-Soft Inc., Tulsa, OK). This software was also used to apply a statistical test for percentage to enable the comparison of N-glycosylation in these groups. P < 0.05 was considered significant.
Results
The adrenal cortex is a major site of novH expression in nontumoral adrenal gland and in human embryos
A Northern blot was performed using RNA extracted from different
human adult endocrine tissues. As shown in Fig. 1A, after a short exposure, novH RNA was
only detected in the adrenal cortex and medulla. However, a very weak
novH expression could be detected in pancreas, thyroid, and stomach
after a longer exposure (not shown). Immunohistochemical experiments
were then carried out to localize the sites of NOVH expression in fetal
and nontumoral postnatal adrenal gland, using the human K19M polyclonal
antibody (24). After 12 wk gestation, the adrenal cortex
consists principally of two distinct zones, the definitive and fetal
zones (see Ref. 38 for a review), both of which contained
NOVH expression (Fig. 1B). At this stage, NOVH expression is more
strongly expressed in the definitive zone, which is composed of a
narrow band of tightly packed basophilic cells (Fig. 1B, c). After 20
wk gestation, NOVH was still detectable in both the definitive and
fetal zones (Fig. 1B, d and e), except in small islands of cells
present in a restricted area of the innermost fetal zone that had a
barely detectable level of NOVH (Fig. 1B, e and f). Immunostaining
subsequently performed on several sections containing this same
restricted zone, using anti-PS100 antibody (Fig. 1B, g and h), which is
specific for peripheral neuronal Schwann cells, and the
antichromogranin A antibody (Fig. 1B, i and j), which is specific for
the chromaffin cells that colonize the medullary part of the adrenal
gland, revealed that the same restricted zone contained cells of
neuroblastic and neuroendocrine origins.
|
, c and d). Thus, during embryogenesis and in nontumoral
adrenal gland, NOVH is mostly expressed in adrenocortical cells. The
apparent discrepancy between the results of the Northern blot and those
of immunohistochemistry is probably due to contaminating adrenal cortex
tissue in the medullary sample. This was confirmed by the detection in
both adrenal cortex and medullary samples of a 1.7-kb mRNA encoded by
the 3ß-hydroxysteroid dehydrogenase (probe provided by Dr. Van
Luu-The) specifically expressed in the adrenal cortex (data not
shown). Normal, benign, and malignant adrenocortical tissues have different NOVH protein profiles
NOVH expression was also studied in both normal and tumoral
adrenocortical extracts (Fig. 2A). The
K19M antibody detected a major 46-kDa band and a faint smear extending
from 46–52 kDa in all normal adult adrenocortical tissues tested. In
benign and suspect tumors from group 1 (Fig. 2A), various amounts of
the 46-kDa form of NOVH were detected, but densitometric analyses of
the NOVH smear extending from 46–52 kDa in 6 of the 11 tumors tested
was significantly (P < 0.05) more intense (mean
± SD, 2.1 ± 1.2) than that in normal
tissues (0.4 ± 0.2). An additional NOVH-related 31- to 32-kDa
doublet was also detected in samples 1, 2, 3, 9, and 11, with different
intensities. Lower molecular mass forms of NOVH, possibly generated by
cleavage of the full-length protein, have been observed in Wilms’
tumors (24) and in the conditioned medium of insect cells
infected with a recombinant baculovirus expressing NOVH
(39).
|
) was
detected. In the malignant tumors in which NOVH expression could be
detected, two different patterns were observed: 1) an intense smear
(no. 21, 25, 28, and 30) as observed in benign tumors with different
intensities of the 46-kDa band; the average intensity of the smear
detected in these four tumors (1.3 ± 0.3) was also significantly
(P < 0.05) higher than that in normal adrenocortical
tissues, but did not differ significantly from that in group 1
(2.1 ± 1.2); or 2) the 46-kDa form present at various levels with
no smear (no. 15, 17, 20, 22, 23, and 29). The 31- to 32-kDa doublet
was detected in various amounts in several malignant tumors (no. 18,
19, 22, 25, 28, and 29), but we could not establish a correlation
between the presence of this NOVH-related form and the stage of the
tumor. It is noteworthy that although classified as benign, tumor 4, in which the IGF-II gene is overexpressed, contained no detectable NOVH.
We found variable amounts of NOVH in tumors from groups 1 and 2, with
overlap with the normal range. However, the total immunoreactivity of
the different forms of NOVH in tumors from group 2 was significantly
less than that in group 1 (Fig. 2B; P < 0.004).
The limited number of normal adrenal tissues available in this study did not allow us to measure significant statistical variations in NOVH levels between normal adrenal gland and tumors from group 1 or 2; however, the general trend is for NOVH levels to be higher in group 1 tumors and lower in group 2 tumors than in normal tissues. Thus, these data strongly suggest that NOVH protein expression is qualitatively and quantitatively altered during adrenocortical tumorigenesis.
Differences in the levels of N-glycosylation of NOVH protein
NOVH protein contains two potential sites of
N-glycosylation, at positions 97 (NQTG) and 280 (NCTS), and
the treatment of novH-transfected MDCK cells with tunicamycin reduces
the apparent molecular mass of NOVH in these cells to 39 kDa
(24). This suggested that the 46- to 52-kDa smear of NOVH
observed in several tumors results from different degrees of
N-glycosylation. We investigated this possibility by
subjecting total protein extract (40 µg) derived from one benign
(tumor 1) and two malignant tumors (tumors 21 and 28) to
N-glycosidase F treatment. The apparent molecular mass of
NOVH under reducing conditions after N-glycosidase F
treatment decreased from 46 to 44 kDa in tumors 1, 21, and 28 (Fig. 3). The forms of NOVH with the lowest
molecular mass detected in samples 1 and 28 were also reduced from
31–32 kDa to the same molecular mass (24 kDa). Therefore, these
results show that during adrenocortical tumorigenesis, changes in NOVH
N-glycosylation can be detected as early as the benign
stage. Serum-free conditioned medium from insect cells infected with a
recombinant baculovirus expressing NOVH (SF9/82) was subjected to the
same conditions and used as a control of N-glycosidase F
treatment. The apparent molecular mass of these NOVH recombinant forms
before N-glycosidase F treatment was lower (44 and 27 kDa)
than that in adrenal tissues, indicative of different levels of
posttranslational modifications. N-Glycosidase F
treatment also reduced NOVH recombinant sizes to 42 and 24 kDa. In our
experimental conditions, however, the apparent molecular masses of the
various forms of NOVH were not reduced to the predicted 39 and 19 kDa,
indicating that either the N-glycosidase F treatment was not
complete or NOVH undergoes other posttranslational modifications in
these samples.
|
As different NOVH-sized proteins due to glycosylation variations
were observed in malignant adrenocortical tumors, we investigated
whether a relationship could be established between these profiles and
the histology of these tumors. For this experiment, we studied benign
tumor 9 and malignant tumors 19, 21, and 28 with a very low level of
NOVH expression or with different electrophoretic profiles, as detected
by Western blotting. Histological examination of tumors 19, 21, and 28
revealed no striking differences. Tumors (no. 19, 21, and 28; Fig. 4B, a, d,
and g) consisted mostly of compact cells grouped in large sheets or
trabeculae separated by a fine fibrovascular stroma. Large areas of
necrosis were also present. Few mitoses were observed in nuclei with
abnormalities.
|
, b–d), NOVH protein was detected
both in the part of the adrenocortical gland that was normal and in the
adenomatous tissue. The level of NOVH expression seemed to be higher in
the adrenal compartment next to the tumoral tissue than in the
adenomatous tissue (Fig. 4A, b). In contrast, little or no NOVH
expression was detected in the capsule composed of connective tissue.
At higher magnification, the clear cells of the adrenal cortex in
normal (Fig. 4A, c) and adenomatous (Fig. 4A, d) tissue expressed NOVH.
In malignant tumors, consistent with Western blot analysis, NOVH was
detected from weak in tumor 19 to high levels in tumors 21 and 28,
essentially in compact tumoral cells (Fig. 4B, b, e, and h). No NOVH
expression was detected in necrotic areas. In Fig. 4B (c, f, and i), a
section is shown at higher magnification to demonstrate that NOVH
detection was not uniform in all tumoral cells. Within the
fibrovascular stroma, some fibroblasts were also positive (Fig. 4B, j).
As previously observed in other tissues (24), NOVH protein
was also present in differentiated structures within the tumor, such as
endothelial cells, and in the surrounding smooth muscle of some blood
vessels (Fig. 4B, k). Transcriptional regulation of novH also occurs in adrenocortical tumors
We investigated whether the differences in the amounts of NOVH
proteins detected in adrenocortical tumors resulted from differences in
levels of novH mRNA by performing Northern blot analyses. Various
amounts of the 2.5-kb novH-specific mRNA species were detected in
different tumors (Fig. 5A), with
significantly more (P < 0.001) in tumors from benign
group 1 tumors than in those from malignant group 2 tumors (Fig. 5B).
There was a correlation between novH protein and RNA levels in the
various samples tested (Fig. 5C; P < 0.001). However,
the ratio of protein to RNA was 6 to more than 100 times higher in
several group 2 tumors such as 15, 17, 20, and 23, than in group 1
tumors 1, 2, and 3, indicating that the level of novH mRNA translation
or NOVH protein stability may be higher in some group 2 tumors. Similar
results were obtained when levels of novH mRNA were normalized relative
to GAPDH (Fig. 5, B and C) or to the ribosomal protein S26 mRNA
(40) (data not shown).
|
Discussion
Several studies (10, 31, 41) have shown that nov expression is widely distributed in normal adult tissues. Levels of nov mRNA have been reported to be high in the brains of adult humans, rats, and chickens, but to differ considerably between species in other organs such as the lung or heart (10, 31, 41). Spatio-temporal regulation of nov expression has also been described in the chicken, because nov mRNA was detected in embryonic, but not adult, heart, muscle, and kidney, whereas in the lung, nov expression was only detected in adult tissue (10). nov expression is tightly regulated during development of the central nervous system (32) and throughout chondrogenesis (Laurent, M., personal communication), suggesting that nov may be involved in the development and differentiation of these tissues.
In this paper we report for the first time that in humans, novH is more strongly expressed in the adrenal gland than in several other adult endocrine tissues. nov was strongly expressed in the adrenal gland cortex derived from the celomic epithelium, and lower levels of its expression were detectable in cells of the medulla of the neural crest origin. NOVH, which is a secreted protein (24), was present in both the definitive and fetal zones of the adrenal cortex in 12- and 20-wk-old fetuses and may therefore play an autocrine/paracrine role in the development and/or differentiation of this tissue.
As novH expression was altered in Wilms’ tumors (20, 24), which have some physiopathological and molecular alterations in common with adrenocortical tumors (2), we investigated whether novH was involved in adrenocortical tumorigenesis by studying its expression in 30 adrenocortical tumors of different types. The major finding of our study is that novH expression is reduced with malignant progression and that different glycosylation patterns can be detected as early as the benign stage.
Both overexpression and down-regulation of nov have been reported in tumors depending on the tumor studied and the histological composition of the tumor. In Wilms’ tumors, high levels of nov expression have been associated with heterotypic tumoral differentiation (e.g. muscle and cartilage) (24). The low levels of novH expression observed in malignant adrenocortical tumors are consistent with the dedifferentiation of tumoral cells and with novH having a potential inhibitory role in the growth of certain types of cells (10). However, immunohistochemical analyses showed that in both benign and malignant tumors, NOVH was not uniformly present in all cells, indicating that novH expression may be dependent on the cell cycle or the stage of differentiation of the cells.
Other members of the CCN family also have different patterns of expression in tumors. The expression of elm1/wisp1 is inversely correlated with the incidence of metastasis and the growth of melanoma cells (16), but is overexpressed in colon tumors. r-Cop1/wisp2 expression is up-regulated in Wnt-1-transformed C57MG cells, but down-regulated in transformed fibroblasts in rats and mice (17). Its level of expression is significantly lower in colon tumors than in normal colon mucosa (15). In breast tumors (15, 42) and in pancreatic tumors (43), the expression of elm1/wisp1, r-Cop1/wisp2, and ctgf has been detected essentially in the stroma cells surrounding tumor cells, with little or no expression in tumor cells. NOVH was detected in adrenocortical tumors, with different amounts in the two compartments. Therefore, although changes in the expression of CCN members are detected in tumors, suggesting a role in tumor growth, no unifying hypothesis has yet been established.
In this study we found that N-glycosylation of the NOVH protein in several benign and malignant adrenocortical tumors differed greatly from that in normal tissue. This modification seems to be specific to NOVH, because no significant alteration of the N-glycosylation profile of other proteins such as IGFBP-3 was detected in the same samples (5) (our data not shown). An increased N-glycosylation of the NOVH protein was significantly (P < 0.05) more frequent in benign (54.5%) than in malignant (22.2%) tumors. In 58% of the benign tumors tested the 46-kDa form of NOVH was also present at higher or similar levels than in normal tissues, whereas in only 11% of the malignant tumors could the level of the 46-kDa band be compared with benign or normal tissues. In a large majority of the malignant samples tested (77%) an important decrease in this form was observed. In several benign and malignant tumors, additional glycosylated 31- to 32-kDa forms of NOVH were observed. No correlation was found between the amounts of these forms and modifications to the N-glycosylation of the 46-kDa form in the various samples studied. This indicates that changes in N-glycosylation do not affect the amount of the lower molecular mass forms of NOVH probably generated by proteolytic cleavage. Changes in NOVH glycosylation may, however, affect the ability of NOVH to interact with other proteins. It has been shown that the extent of N-glycosylation can modulate the cell-binding activity of IGFBP-3 (44). Aberrant N-glycosylation of NOVH could also affect NOVH stability, as suggested by our previous results (24). Along this line, it has been reported that aberrant N-glycosylation of von Willebrand factor type C (45) leads to an increase in clearance from plasma and accounts for a low von Willebrand factor type C phenotype. Further studies are required to elucidate the role of N-glycosylation in the biochemical properties of NOVH.
Variable amounts of NOVH were observed within group 1 and 2 tumors.
However, these variations could not be strictly correlated to any of
the clinical or molecular data indicated in Table 1 in particular. They
are more likely to be due to a combination of several different
elements characterizing the tumors, including the hormonal pattern,
size of the tumor, presurgery treatment, and variable amounts of
IGF-II.
A growing body of evidence suggests that the IGF system is involved in tumor proliferation (reviewed in Ref. 46) and particularly in the development of adrenocortical tumors (2, 3, 4, 5, 6, 47). In this study no overexpression of the IGF-II gene was detected in tumors from group 1, except for tumor 4, which did not express novH, whereas 16 of the 18 malignant tumors overexpressed the IGF-II gene. Twelve of these 16 tumors displayed a strong down-regulation of novH expression. Therefore, although no strict inverse correlation can be drawn, these results suggest that IGF-II and novH may be regulated in opposite ways by a common transcription factor. In some Wilms’ tumors, we have previously reported that levels of novH mRNA were inversely correlated to levels of wt1 mRNA (20) and that wt1 indirectly regulated novH expression ex vivo (25). wt1 also acts as a negative transcriptional regulator for IGF-II (30), and IGF-II gene expression is also altered in Wilms’ tumors (26, 27, 28). wt1 can up-regulate or down-regulate gene expression depending on the interacting proteins and gene promoters (48, 49, 50, 51). It has recently been shown that wt1 expression is required for development of the adrenal cortex (52), but nothing is known about its possible involvement in IGF-II and novH regulation in this tissue. It is also possible that IGF-II and/or novH each regulate the other’s expression.
In conclusion, the significant differences in the levels of novH protein and RNA between benign and malignant tumors may be of importance. For many genes such as IGFII, IGFBP2 (5), p57Kip2, G1 cyclins, and G1 CDKs (7), altered expression has only been observed in malignant tumors. In contrast, differences in novH expression between benign tumors and normal tissue can be detected, suggesting that novH could participate in the early stages of tumorigenesis.
Relatively little is known to date about the function of novH. In several cell systems nov has been reported (53, 54) to be associated with cell quiescence. Together with previous observations that overexpression of nov in chicken embryo fibroblasts led to an inhibition of growth (10), this suggests that nov is a negative regulator of growth. However, different cells may respond differently to NOV, because recombinant NOV stimulates 3T3 cell proliferation (41), but has no effect on the proliferation of vascular smooth muscular cells (VSMC) (54).
Several lines of evidence suggest that nov is more likely to be involved in cell adhesion. The multidomain structure of NOV and the other CCN members suggests that they bind to components of the extracellular matrix, including heparin-like oligomers (14). The finding that fibulin 1C, an extracellular matrix-associated protein (55, 56), interacts with the NOVH protein (39) provides a clue for the possible participation of NOVH in signaling pathways involving the extracellular matrix. More recently, it has been shown that the recombinant NOV protein can promote cell adhesion ex vivo and that changes in nov expression occur in response to injury of the arterial walls (54). It has been proposed that reduced nov expression is involved in releasing VSMC for migration and proliferation and that nov can be reexpressed during the late stages of repair, when migration and proliferation slow down. It is thus tempting to speculate that the enhanced expression of nov in adrenocortical tumors participates in the benign phenotype by increasing adhesion of cells, and that decreased levels of NOV in malignant tumors are involved in cell invasiveness. Alternatively, novH could act as a tumor suppressor in adrenocortical tumors. Further investigation of the biological properties of NOVH in adrenocortical cells will enable us to determine whether changes in the expression of novH play a key role in the development of these tumors.
Note Added in Proof
While this manuscript was submitted, tight regulation of nov expression during mouse development in skeletal and visceral muscles and in nervous system was described (Natarajan D, Andermarcher E, Schofield PN, Boulter C 2000 Mouse Nov gene is expressed in hypaxial musculature and cranial structures derived from neural crest cells and placodes. Dev Dyn 219:417–425).
Acknowledgments
We thank Prof. B. Perbal for helpful discussions and Dr. X. Bertagna and Dr. H. Kleinman for critical reading of this manuscript. We are grateful to A. M. Henere for skillful assistance with the immunohistochemistry experiments, to S. Kyurkchiev for purifying the K19M antibody, and to J. M. Ricort for performing Western ligand blotting for IGFBP-3 detection in adrenocortical samples. We thank Dr. M. Tantau for provision of fetal material. We also thank Mr. J. Grellier for assistance with photography.
Footnotes
This work was supported by Assistance Publique des Hopitaux de Paris (Contrat de Recherche Clinique 97133), University Paris VI, Faculté Saint-Antoine (UPRES EA 1531), Association de Recherche contre le Cancer (no. 1364), Centre National de la Recherche Scientifique, INSERM (U-515), and Programmes Hospitaliers de Recherches Cliniques Grant AOM95201 for the Comete Network.
Abbreviations: GAPDH, Glyceraldehyde-3-phosphate dehydrogenase; IGFBP-2, IGF-binding protein-2; poly(A)+, polyadenylated; VSMC, vascular smooth muscular cells.
Received November 17, 2000.
Accepted April 4, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  S. E Ghayad, J. A Vendrell, I. Bieche, F. Spyratos, C. Dumontet, I. Treilleux, R. Lidereau, and P. A Cohen Identification of TACC1, NOV, and PTTG1 as new candidate genes associated with endocrine therapy resistance in breast cancer J. Mol. Endocrinol., February 1, 2009; 42(2): 87 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Doghman, M. Arhatte, H. Thibout, G. Rodrigues, J. De Moura, S. Grosso, A. N. West, M. Laurent, J.-C. Mas, A. Bongain, et al. Nephroblastoma Overexpressed/Cysteine-Rich Protein 61/Connective Tissue Growth Factor/Nephroblastoma Overexpressed Gene-3 (NOV/CCN3), a Selective Adrenocortical Cell Proapoptotic Factor, Is Down-Regulated in Childhood Adrenocortical Tumors J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3253 - 3260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. N. West, G. A. Neale, S. Pounds, B. C. Figueredo, C. Rodriguez Galindo, M. A. D. Pianovski, A. G. Oliveira Filho, D. Malkin, E. Lalli, R. Ribeiro, et al. Gene Expression Profiling of Childhood Adrenocortical Tumors Cancer Res., January 15, 2007; 67(2): 600 - 608. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. Lin, C.-C. Chen, S.-J. Leu, T. M. Grzeszkiewicz, and L. F. Lau Integrin-dependent Functions of the Angiogenic Inducer NOV (CCN3): IMPLICATION IN WOUND HEALING J. Biol. Chem., March 4, 2005; 280(9): 8229 - 8237. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. G Jiang, G. Watkins, O. Fodstad, A. Douglas-Jones, K. Mokbel, and R. E Mansel Differential expression of the CCN family members Cyr61, CTGF and Nov in human breast cancer Endocr. Relat. Cancer, December 1, 2004; 11(4): 781 - 791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. T. Fu, J. F. Bechberger, M. A. Ozog, B. Perbal, and C. C. Naus CCN3 (NOV) Interacts with Connexin43 in C6 Glioma Cells: POSSIBLE MECHANISM OF CONNEXIN-MEDIATED GROWTH SUPPRESSION J. Biol. Chem., August 27, 2004; 279(35): 36943 - 36950. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C A Longui, S H V Lemos-Marini, B Figueiredo, B B Mendonca, M Castro, R Liberatore Jr, C Watanabe, C L P Lancellotti, M N Rocha, M B Melo, et al. Inhibin {alpha}-subunit (INHA) gene and locus changes in paediatric adrenocortical tumours from TP53 R337H mutation heterozygote carriers J. Med. Genet., May 1, 2004; 41(5): 354 - 359. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. R. GROTENDORST, H. RAHMANIE, and M. R. DUNCAN Combinatorial signaling pathways determine fibroblast proliferation and myofibroblast differentiation FASEB J, March 1, 2004; 18(3): 469 - 479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bertherat, L. Groussin, F. Sandrini, L. Matyakhina, T. Bei, S. Stergiopoulos, T. Papageorgiou, I. Bourdeau, L. S. Kirschner, C. Vincent-Dejean, et al. Molecular and Functional Analysis of PRKAR1A and its Locus (17q22-24) in Sporadic Adrenocortical Tumors: 17q Losses, Somatic Mutations, and Protein Kinase A Expression and Activity Cancer Res., September 1, 2003; 63(17): 5308 - 5319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. O. Muench, J. V. Ratcliffe, M. Nakanishi, H. Ishimoto, and R. B. Jaffe Isolation of Definitive Zone and Chromaffin Cells Based upon Expression of CD56 (Neural Cell Adhesion Molecule) in the Human Fetal Adrenal Gland J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3921 - 3930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ratcliffe, M. Nakanishi, and R. B. Jaffe Identification of Definitive and Fetal Zone Markers in the Human Fetal Adrenal Gland Reveals Putative Developmental Genes J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3272 - 3277. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. G. Lin, S.-J. Leu, N. Chen, C. M. Tebeau, S.-X. Lin, C.-Y. Yeung, and L. F. Lau CCN3 (NOV) Is a Novel Angiogenic Regulator of the CCN Protein Family J. Biol. Chem., June 20, 2003; 278(26): 24200 - 24208. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Thibout, C. Martinerie, C. Creminon, F. Godeau, P. Boudou, Y. Le Bouc, and M. Laurent Characterization of Human NOV in Biological Fluids: An Enzyme Immunoassay for the Quantification of Human NOV in Sera from Patients with Diseases of the Adrenal Gland and of the Nervous System J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 327 - 336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Lafont, M. Laurent, H. Thibout, F. Lallemand, Y. Le Bouc, A. Atfi, and C. Martinerie The Expression of novH in Adrenocortical Cells Is Down-regulated by TGFbeta 1 through c-Jun in a Smad-independent Manner J. Biol. Chem., October 18, 2002; 277(43): 41220 - 41229. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
